thomas leverett weblog

thank you for your cupertino

OK so it started with that first article: Mr. Zimmer says that Cupertinos, which are wrong words offered by your spell-checker, have been around for a while, are increasing due to the frequency of mobile devices, and have the capacity to be very interesting. It also comes out in digging up the history, that the name Cupertino, though a city in California that is ironically the home of Apple, has been the name for these (I have called them spell-check absurdities) since long ago, when in Europe, early spell-checks did not have the unhyphenated cooperation and could only change it to Cupertino...read it and weep.

The series of articles brings up a number of points, besides his, which I generally agree with. I in fact have been studying this phenomenon for a while with a couple of added twists: that spell-check was not made for international learners, and in fact has trouble with their irregular data; that spell-check and grammar-check influence not only their writing but also their learning, that there is such a thing as a grammar cupertino (you read it here first)...and finally that the changing world of correction algorithms will influence the world for years to come but will probably help our students last, since their errors are so uncommon yet at the same time systematic and egregious.

Let's take the last one first. Granted, their algorithms are more sophisticated than they used to be: that's why it's hard for us to believe, today, that a computer could mistake Cupertino for cooperation. One of the best examples is when it changes definately to defiantly because the algorithm says, reverse the letters first, before you assume that they got the wrong vowel. But that algorithm can be changed if, 99% of the time, they got the wrong vowel. And their data shows that that is in fact what is happening. I won't go into why what our students type in is so different from what your average American poor-speller types in. Therein, however, lies the rub.

Here's a grammar cupertino. Student types in "We like each others." Grammar-check checks with its loads of data, and comes back and says, 99% of times when each other has an -s, it has an apostrophe. Student changes it, on advice, to "We like each other's"...We now have a grammatical structure that is legitimate (in its own right) - but absolutely in the wrong place, and not what the student means. It creates an error the teacher has no idea how to deal with (What does he mean? Where did the apostrophe come from?) - and one that would never have been generated naturally by other means.

Why do I refuse to capitalize cupertino? Well, if I ever go to the city, I'll capitalize it. But I think, if we're talking about words, and it's a word, we're better off if we just bring it down to my level right from the start. I'd never heard the word until this weekend, but I was delighted to find it, because, as I've said, it's up my alley in terms of what I study. Which reminds me, a whole vacation I spent restoring my own work - dozens of articles, some of them antiquated; nonetheless, if you're interested, read 'em & weep. The newest, Negotiate with the elephant, is on this very topic; I just wrote it, what, maybe Tuesday. My wife suspected that it was about living with someone who wouldn't lose weight, but I disabused her of that misperception.

Zimmer, B (2011, Jan. 13). Auto (in)correct. New York Times Magazine. http://www.nytimes.com/2011/01/16/magazine/16FOB-onlanguage-t.html?_r=1. Accessed 1-11.

Zimmer, B. (2006, Oct. 2). The Cupertino effect strikes again. Language Log. http://itre.cis.upenn.edu/~myl/languagelog/archives/003629.html. Accessed 1-11.

Zimmer, B. (2007). When Spellcheckers Attack: Perils of the Cupertino effect. OUPblog (Oxford University Press): From A to Zimmer. http://blog.oup.com/2007/11/spellchecker/. Accessed 1-11.

Labels: grammar

posted by tom @ 5:08 PM